metadata
dataset_info:
features:
- name: image
dtype: image
- name: id
dtype: int64
- name: embeddings
sequence: float32
splits:
- name: train
num_bytes: 25739969
num_examples: 67
download_size: 25845836
dataset_size: 25739969
configs:
- config_name: default
data_files:
- split: train
path: data/train-*